امنیت کامپیوتر برای فن آوری های جمع آوری داده ها Computer security for data collection technologies

1. Introduction Technology has become an important tool for many non-governmental organizations (NGOs) and groups collecting data in the developing world. For example, technology can provide people in remote regions with access to financial services and allow organizations to collect vital information within communities they serve. Information and Communication Technology for Development (ICTD) is the study of what technology can accomplish and how technology is used in such low-resource settings around the world. ICTD takes a broad definition of “low-resource”. Areas affected by poverty are frequently the focus of ICTD, but any setting where things like limited connectivity, unreliable power, or lightly skilled personnel conspire to create a unique technological landscape might be relevant to ICTD. Although there have been some efforts to study and address computer security and privacy risks with technologies in an ICTD environment, both on a case-by-case basis for specific technologies and from an academic perspective, e.g., [1, 9, 26], the space of “computer security meets ICTD” is still in its infancy. We contribute to this space through insights into how to evaluate and address computer security risks in ICTD environments. To provide a foundation for our insights, we choose to focus on a particular class of technologies — data collection toolkits— and, in particular, a specific, widely-used instance of such a technology: Open Data Kit (ODK). Data is crucial for many NGOs and researchers to monitor and evaluate deployments or interventions and report to donors on activities. For example, organizations might collect patient information during clinic visits, assess the prevalence of pests in rural farmland, or document infrastructure in need of repair. ODK allows digital forms to be created without deep technical expertise, and has been used as a platform by numerous organizations. By studying computer security risks with ODK, we are able to extract lessons for both ODK and other data collection deployments, as well as infer lessons for other new ICTD technologies.